التنبؤ بمخاطرة أسعار الصرف باستخدام الشبكة العصبونية : دراسة تطبيقية

Abstract EN

Neural Networks technology is characterized by its ability to process inputs and reality data and bypassing traditional techniques in explaining the behavior of phenomena, which has made it a topic of interest to many researchers in different fields.

The research aims to take advantage of the distinct capabilities of the neural networks when using it in making investment decisions, especially in the field of currencies, as it is one of the important areas of investment, whether at the level of individuals or financial institutions with the aim of achieving profits, which depends mainly on predicting the movements of currency prices up or down.

Three types of neural networks technologies were used: Clustering Modified Radial Basis Function (MRBF), (Multi-layer perception (MLP) technology and finally (Artificial neural network (ANN)) technology to obtain the search results.

The analysis was carried out using monthly information on the exchange rate of the euro against the US dollar for the period from 1/1/2015 to 31/12/2020, and MATLAB version V.5.2.0 was used to process the input data and thus obtain the search results.

The research found that the use of a neural network (ANN) gives great accuracy in predicting the future movements of the currency of the research sample compared to other techniques used according to the RMSE comparison standard, which reduces the risk of investing in currencies and thus increases returns.